1. Field of the Invention
The invention relates to a method of imaging perfusion parameters of a liquid in a first part of a body by means of magnetic resonance, which method includes the following steps: measuring MR signals of spins in the first part by generating MR imaging pulse sequences, reconstructing MR images from the measured MR signals, determining an image containing the perfusion parameters of the first part from a combination of the reconstructed MR images and an input function of a contrast medium which flows to the first part of the body via a supply for the liquid, and determining, using magnetic resonance, information for determining the input function of the contrast medium.
2. Description of Related Art
The invention also relates to an MR device for carrying out said method. In the context of the present patent application, a k-space is to be understood to mean a spatial frequency domain in which a path is followed during the measurement of the MR signals by application of gradients to the steady magnetic field. The path in the k-space is determined by the time integral of the gradients applied during the interval between the excitation of the spins and the instant in time at which the MR signal is measured. The measured values of the MR signals which correspond to the most important part of the path or paths yield the inverse Fourier transformed values of an image of the object. Furthermore, a gradient is to be understood to mean a temporary magnetic field which is superposed on a steady magnetic field and causes a gradient in the steady magnetic field.
The above method is known from the article xe2x80x9cHigh Resolution Measurement of Cerebral Blood Flow Using Intravascular Tracer Bolus Passages. Part II: Experimental Comparison and Preliminary Resultsxe2x80x9d as published by L. Ostergard et al. in Magnetic Resonance in Medicine No. 36, 1996, pp. 726-736. The known method is used to determine, for example perfusion parameters of the liquid in the first part of the body, for example a cerebral blood flow in a part of the brain. Perfusion parameters indicate an exchange between inter-cellular and extra cellular fluid. The perfusion parameters can be represented in an MR perfusion image which may be used as an aid to a physician to detect early stroke in the brain of the body to be examined. Furthermore, according to the known method MR signals are measured in the part of the brain by means of EPI pulse sequences in order to reconstruct a first series of MR images so as to trace the progression in time of the cerebral blood flow. In order to obtain a quantitative distribution of a cerebral blood flow in the part of the brain. The perfusion MR image is determined from the combination of the MR-image and the input function of the contrast medium. The input function represents a concentration of the contrast medium in the supply as a function of time. The MR image is deconvoluted with the input function of the contrast medium in a supply to the part of the brain. An example of such an input function comprises the arterial input function of the middle cerebral artery. In order to determine the arterial input function of the middle cerebral artery, the known method utilizes EPI MR imaging sequences to measure MR signals for a second series of MR images of a second part of the body which comprises a part of the blood supply to the part of the brain. After reconstruction of the second MR images, an operator determines a number of, for example from 5 to 10 arterial pixels in the second MR images, which arterial pixels represent a part of the middle cerebral artery. The arterial input function is determined on the basis of the variation in time of the intensity of the arterial pixels. In order to obtain, for example an image of a cerebral blood flow of voxels of the part of the brain, pixels in the series of MR images which correspond to the voxels are deconvoluted with the arterial input function obtained. It is a drawback of the known method that the determination of the arterial input function from the arterial pixels is not reproducible and that errors are liable to occur in the determination of the arterial input function.
It is an object of the invention to reduce errors in the determination of the arterial input function. To this end, the method according to the invention is characterized in that the input function is determined by application of a input pulse sequence for the excitation of spins in a second elongated part of the body and measuring input function MR-signal from the second part, the second elongated part of the body comprising a part of a vessel transporting the contrast medium to the first part, a cross section of the second elongated part being chosen such that a contribution from excited spins from the contrast medium in the vessel in the second elongated part dominates a contribution from other excited spins of the second elongated part to the MR-signal. For example, when the cross-section of the elongated second part equals twice the diameter of the vessel the contribution from the excited spins of the contrast medium in the part of the vessel dominates the contribution from the excited other spins from the part of the second elongated part. Because the input pulse sequence can be a very short sequence compared to the standard MR-imaging sequences a high temporal resolution of the input function can be obtained. Furthermore, the input function can be reproducibly determined from the MR-signal. In order to determine an input function the measured MR-signal is fourier transformed and the spin densities corresponding to the different positions in the second elongated part can be obtained from the fourier transformed MR-signals. The input function can be obtained, for example, by summing the spin densities for all the different positions.
A special version of the method according to the invention is characterized in that in order to measure the input function MR signals, the input pulse sequence includes a measuring gradient which is oriented along a longitudinal axis of the elongate second part. Thus, the input function MR signal is measured wherefrom a transverse magnetization of a position along the gradient can be determined by means of a 1D Fourier transformation. An advantage of the use of an input pulse sequence containing only a single measuring gradient resides in the fact that the input function MR signal can be quickly measured relative to a bolus passage of the contrast medium, so that the accuracy of the input function is enhanced. By fast measurement of the input function MR signal relative to the local measuring time of MR signals for reconstructing a single image of the series of MR images, disturbances in the magnetization in the first part during the measurement of the MR signals for the reconstruction of the series of images are minimized, so that the risk of artefacts in the images is reduced.
A further version of the method according to the invention is characterized in that for the determination of the input function from the measured input function MR signals a value of the input function associated with an instant at which an input function MR signal has been measured is determined from a combination of elements of a one-dimensional Fourier transformation of the input function MR signal. For example, summing the amplitudes of the one-dimensional Fourier transformation of the input function MR signal enables determination of a value which is related to a concentration of the contrast medium present in the part of the supply at the instant at which the input function MR signal has been measured.
A further version of the method according to the invention is characterized in that the input pulse sequence includes a 2D excitation RF pulse for excitation of the spins in the second elongate part. The 2D excitation RF pulse is known from the article xe2x80x9cA k-space analysis of small tip angle excitationxe2x80x9d, by J. Pauly et al. as published in Journal of Magnetic Resonance 1989, No. 81, pp. 43-56. When use is made of a 2D excitation RF pulse, an elongate second part which comprises a cylinder can be excited in a short period of time, for example 6 ms, so that the measuring time of the input function MR signal is further reduced and the accuracy of the input function is enhanced.
A further version of the method according to the invention is characterized in that the MR imaging pulse sequences include an echo planar imaging (=EPI) imaging pulse sequence. The use of EPI imaging pulse sequences enables fast measurement of the MR signals required for determining the MR image. An example of such a known EPI imaging pulse sequence is a multislice EPI imaging pulse sequence or a 3D EPI imaging pulse sequence.
A further version of the method according to the invention is characterized in that in the EPI MR imaging pulse sequence an echo time Te between an excitation RF pulse and a corresponding MR signal is greater than a repetition time Tr between two temporally successive excitation RF pulses. As a result of this step, the MR signals are obtained with a small range of T2* weights, where T2* represents an actual transverse magnetic time constant. Such an EPI imaging pulse sequence is known from the article xe2x80x9cA Functional MRI Technique Combining Principles of Echo-Shifting with a Train of Observations (PRESTO)xe2x80x9d as published by G. Liu et al. in xe2x80x9cMagnetic Resonance in Medicinexe2x80x9d, No. 30, 1993, pp. 764-768. A further version of the method according to the invention is characterized in that the elongate second part is chosen to extend transversely of the direction of the supply. The accuracy of the input function can be further enhanced by choosing the elongate second part to extend transversely of the supply.
The invention also relates to an MR device.